In order to assess the number of probes that, nevertheless, report genome-specific expression, we used the presence or absence of significant signal (qualitative expression) for each probe to classify the expression pattern of the corresponding unigene.
Despite these limitations, these data can be used to classify the expression patterns of the leaf proteases.
As presented in Table 1, the median value was set as the cutoff to classify the expression level of Linc00974 while KRT19 was divided into positive and negative groups according to IHC staining.
Thus, the cut-off value was set at 1.0 and was used to classify the tumoral expression of sIL-6 as over-expression or under-expression compared with normal mucosa.
Laukka et al. [ 16] employed machine learning simulations to classify the vocal expression of emotions produced by professional actors in 5 English-speaking cultures.
We used data from in situ hybridization experiments that classified the expression of about 300 genes as FC specific, FCM specific or both (Estrada et al., 2006).
We used hierarchical clustering to classify the changes in expression of miRNAs that were significantly differentially expressed in Array A and Array B. The miRNA expression profiles of SKOV3 and SKOV3/DDP cells were confirmed with miRNA-specific quantitative real-time RT-PCR.
The percentage of pSTAT3 nuclear positive cells were used to classify the grade of its expression as negative (< 5%), weak (6-50%), and strong (51-100%).
Venn diagrams were used to classify the total number of genes whose expression change was specific to or common in the comparisons of CDV treatment for 24 h, 48 h and 72 h.
For example, an analysis of yeast strains grown in different nutritional conditions combined Z-scores of metabolic fluxes obtained by either metabolic or gene expression measurements to classify the regulation level of metabolic circuits as transcriptionally, post-transcriptionally, or metabolically controlled [ 46].
